The Use and Reuse of Stone Circles

by Richard Bradley

  A number of structural elements at Laikenbuie recall the characteristics of other monuments in the region. The most striking elements are the use of quartz, the organisation of coloured stones in the kerb, and the striking configuration of the southern perimeter of the ring cairn. Quartz was widely used in Northern British monuments and its presence at Laikenbuie confirms that this practice continued into the Early Iron Age, but in this case the monument included rounded boulders which had remained intact – broken pieces like those described in Chapters 2–5 were not represented at Laikenbuie.

  More striking is the selection of differently-coloured stones for incorporation in the kerb, and particularly the important distinction between the distributions of red and white boulders in the ring cairn. A striking parallel in the same region is the kerb cairn at Balnuaran of Clava with its red ‘threshold’ and alternating sections of pink and white stones (Bradley 2000, 38–39). That particular structure echoes earlier monuments on the same site because the heights of those kerbstones are graded towards the southwest. The monument at Laikenbuie has a comparable feature, for on its southern perimeter it includes a low setting of orthostats flanked by two taller uprights.

  The history of these distinctive structures is considered further in Chapter 8.

  PART TWO

  The Excavated Monuments in their Wider Contexts

  CHAPTER SEVEN

  After the Great Stone Circles

  Richard Bradley

  The first stone circles

  Colin Richards’s (2013) book Building the Great Stone Circles of the North investigates a series of Neolithic monuments, in particular those in Orkney and the Western Isles. As his title makes clear, they were substantial structures. They were built of unusually tall monoliths and some of them enclosed considerable areas. Aubrey Burl (2000, 33–38) compares these monuments with the large henges of the mid-third millennium BC and suggests that they were constructed and used by similar numbers of people.

  Until comparatively recently the antiquity of the first stone circles had been difficult for prehistorians to accommodate. They followed the same assumptions as Piggott and Simpson (1971) in their report on Croftmoraig. Timber circles seemed to be older than those built of stone because there were sites where a wooden structure was replaced by a setting of monoliths with the same footprint. That was probably the case with The Sanctuary at Avebury (Pollard 1992), and in Scotland a similar sequence has been identified at Temple Wood (Scott 1989) and on Machrie Moor (Haggarty 1991).

  New projects have undermined this scheme. There are absolute dates from a number of stone circles which show that they originated around 3000/2900 BC. Well-known examples include the Stones of Stenness, Balbirnie, Calanais and the smaller of two neighbouring structures at Temple Wood (Sheridan 2004a; Gibson 2010a). Similar dates can be suggested on the basis of artefact associations, including the monument at Ballynoe in Ireland (Groenman-van-Waateringe and Butler 1976). An outlying monolith at Long Meg and her Daughters in Cumbria is decorated in a style related to passage grave art (Bradley 2007, 120–21). The same is true of the larger structure at Temple Wood (Scott 1989).

  The chronology of timber circles has changed in parallel with new studies of stone monuments. Among the earliest examples is the Scottish site at Carsie Mains, where one such structure was built around 3000 BC (Brophy and Barclay 2004). In eastern England, another seems to have occupied the terminal of the Springfield Cursus where it was associated with Middle Neolithic pottery (Buckley et al. 2001). If these buildings were contemporary with the first stone circles, the last freestanding rings of posts were built almost two thousand years later. New evidence from the Kilmartin complex is particularly revealing, for in one case a timber circle was replaced by a stone circle during the Neolithic period. On another site not far away there were two more rings of posts dating from the Early and Middle Bronze Ages respectively (Cook et al. 2010).

  The sizes of timber circles changed over time. Alex Gibson (2003) has shown that the earliest examples were comparatively small. They became significantly larger at the transition between the Late Neolithic and Chalcolithic periods, and then their scale decreased. Some of these buildings are associated with radiocarbon dates. That is important as most stone circles are of uncertain age. It seems possible that their dimensions followed a similar trajectory. The largest of all are the Great Stone Circles discussed by Colin Richards.

  Recumbent stone circles, Clava Cairns and other monuments

  The monuments investigated in this chapter are less extensive and were later in date than most of those considered in Richards’s book. There is no evidence that they had been used during the Neolithic period. Investigations of Clava Cairns and recumbent stone circles came to a similar conclusion. The same was true of the Aberdeenshire henge monument at Broomend of Crichie which is in the same area as several settings of monoliths. Far from being built during the Neolithic period as had been expected, it actually dated from the Bronze Age (Bradley 2011, chapters 1–3).

  Perhaps the most revealing of these structures are the Clava Cairns which were investigated between 1994 and 1997, for they illustrate the ideas that were current twenty years ago. The monuments consist of ring cairns and passage graves, each of them enclosed within a circle of monoliths. They had been dated to the Middle or Late Neolithic, but this was based on two unwarranted assumptions. The first was that all chambered tombs were Neolithic, and the second that those at Balnuaran were contemporary with a superficially similar monument at Newgrange in Ireland. New fieldwork at the Scottish sites established that they were built during the Chalcolithic period or even the Early Bronze Age. In the same way, a new interpretation of Newgrange suggests that the stone circle there was not part of the original monument (Stout and Stout 2008, 84–92).

  In fact the Clava Cairns were only one kind of chambered tomb dating from the late third millennium BC and the beginning of the second. A programme of excavation and radiocarbon dating has shown that the same applies to the Irish monuments known as wedge tombs (O’Brien 1999). It seems possible that other groups of small chambered tombs were built at about the same time. They include the entrance graves of Cornwall and the Scilly Islands and other structures in south-west Scotland (Jones and Thomas 2010). Individual sites conform to the same broad pattern, and excavation at Calanais in the Hebrides has established that a chambered cairn erected inside an older stone circle was associated with Beaker pottery (Ashmore 2016).

  Many of these examples share a common characteristic. In contrast to a number of older monuments they are not directed towards the rising sun – they are orientated towards the south and, especially, to the southwest (Bradley 2016). The Clava passage graves are enclosed by rings of standing stones, the tallest of which were placed in front of the entrance. Exactly the same happens at circles like Tomnaverie, where the highest monoliths framed the position of a recumbent stone, and the lowest were on the opposite side of the circuit. At times this practice extended to other parts of these structures, so that the kerbstones at both kinds of monument might follow the same convention. In certain cases particular uprights were matched with those around the edge of the cairn by selecting raw material which shared the same colours, shapes or textures.

  Figure 7.1. Two chambered cairns with south-western alignments, external platforms and stone circles. Balnuaran of Clava northeast cairn is in northern Scotland and the wedge tomb at Clogherny Meenerrigal is in Ireland. Information from Bradley 2009.

  There are some striking contrasts between the monuments on either side of the Irish Sea (Fig. 7.1). Like older passage graves, wedge tombs contain deposits of cremated human bone, but the burnt bone found in the Scottish Clava Cairns may result from later reuse, as Piggott’s excavation at Corrimony found that its chamber contained an inhumation burial (Piggott 1956, 180–83). Even so, there are a few monuments which suggest direct connections between these regions. For instance, a large cairn at Kintraw on the west coast of Scotland has a graded kerb exactly like a C
lava Cairn and there is even a stone setting on its south-west side which resembles the focal point of a recumbent stone circle (Simpson 1967). Similarly, at the Irish site of Beltany Tops a ring of stones enclosed what seems to have been a platform. The monument incorporated an alignment from northeast to southwest marked by the changing heights of the monoliths and the placing of cup marks (Lacey 1983, 72–73). A stone circle enclosed the wedge tomb at Clogherny Meenerrigal (Davies 1939), and the same may have happened at Island where the setting of monoliths dated from the Middle Bronze Age (Bradley 2009, 227).

  The Great Stone Circles and their successors

  Taken together, these features suggest some ways of distinguishing the later stone circles from those considered by Richards.

  Their forms are very different. Although the first stone circles could be of various sizes, most were defined by tall monoliths. This is a characteristic of sites like Calanais and the Stones of Stenness (Sheridan 2004a). With the doubtful exception of Balbirnie, there is no evidence that the pillars were graded by height, nor is there anything to suggest a special emphasis on the southwest. The first major monument where it is found is the sarsen horseshoe at Stonehenge, which was not the earliest structure on the site. It was built towards the middle of the third millennium BC, several centuries after Calanais, Stenness and Temple Wood (Darvill et al. 2012).

  Nor do the Great Stone Circles show a similar sequence to the excavated monuments considered in the first part of this book. The recumbent stone circle at Hillhead was built around an already existing cairn, and the same happened at Tomnaverie and Cothiemuir Wood and probably on other sites. The Clava Cairns illustrate a similar sequence, for at Balnuaran the settings of monoliths were not constructed until a ring cairn and two passage graves were already in place. The interval may have been brief, but in every instance it suggests that the raising of the standing stones brought the structural sequence to an end.

  It is not clear that the same applied to the Great Stone Circles. At Calanais a clay platform was built inside the already-existing monument and a small building was erected on its surface (Ashmore in press). Richards suggests that something similar may have happened at the Stones of Stenness, but here the platform is much larger and the evidence is less conclusive (2013, 68–78). What is clear is that the internal structures at Calanais were replaced by a small passage grave which could have belonged to the same generation of megalith building as Clava Cairns and wedge tombs. If that is right, the monuments studied by Richards provided the setting for further structures inside them. In contrast, stone circles like those at Tomnaverie and Balnuaran of Clava represent the final development on individual sites and may have closed access to the interior. Of course, that development did not last, and after an interval most of the monuments were reused. That practice is discussed in Chapter 8.

  A last observation concerns the distinctive character of stone circles in Northern Britain compared with those in the south. So far this discussion has contrasted monuments which provided a setting for later buildings with the sequence at Chalcolithic and Bronze Age sites where a ring of monoliths enclosed an older structure and brought its use to an end. The kind of sequence identified at places like Clava and Tomnaverie is particularly common in Scotland. It is less often found in the south where most circles seem to have been empty.

  Building the Great Stone Circles

  Richards (2013) suggests that the building of the Great Stone Circles played an important role in Neolithic society. This is shown in two ways.

  The first is the enormous effort that went into quarrying, moving and erecting the stones. It was a dangerous activity and one which might have involved large numbers of people. Studies which apply modern engineering methods to the past miss the point that the creation of Great Stone Circles would have been a special event witnessed by a considerable audience. The work played a fundamental part in the dynamics of ancient communities. The demanding nature of the task also made it a medium for competition between different social groups. Rather than emphasising the final form of any structure, Richards focuses on the processes by which individual stones were selected, moved and deployed. It follows that a single monument might have been constructed over a lengthy period and that some, like the Stones of Stenness, were never completed at all. For Richards, the very notion of a ‘completed’ structure is problematical. The process of assembling these buildings may have been far more important than their projected form, and there was always the possibility of accidents and changes of plan before the task was accomplished. An important alternative, illustrated by structures on the island of Lewis, was to use materials that were readily to hand and to prop them in position to create an effect that was unlikely to last a long time. Richards calls this ‘expedient monumentality’ (2013, chapter 9).

  Richards’s second observation is related to all these points. On the Mainland of Orkney it is clear that the two stone circles – those at Stenness and Brodgar – were composite structures formed out of raw materials obtained from several different sources. The task of bringing these materials from separate places – one of them the site of a chambered tomb – meant that the monuments could stand for the wider community as well as the groups of people who extracted and transported the stones. In this sense Stenness and Brodgar provided microcosms of the island as a whole. Such evidence is not restricted to Scottish sites, or to one kind of monument. As early as 1887 Lloyd Morgan recognised the same process at the Stanton Drew stone circles in Somerset which were constructed out of raw materials from a variety of geological deposits. It is also true of Stonehenge (Parker Pearson 2012). The same principle was expressed by the passage grave of Newgrange. Although it had a long history, its construction and use combined materials obtained along much of the eastern seaboard of Ireland (Eogan 1999).

  Similarities and contrasts

  The monuments considered in Part 1 are very different from the Great Stone Circles characterised by Richards. Not only were they built at a later time, they had a completely different emphasis.

  The most striking contrasts are between the Great Stone Circles and both recumbent stone circles and Clava Cairns. They are emphasised here as these structures were closer in time than the other monuments discussed in this chapter. In several respects they possess an architectural unity that sets them apart from earlier buildings.

  Two of the features shared by recumbent stone circles and Clava Cairns are the grading of monoliths by height and an orientation between the south and southwest. The changing heights of the stones emphasised the preferred direction, with the lowest on one side of the circle and the tallest directly opposite. At Clava passage graves the focal point was the entrance to the chambered tomb, and in recumbent stone circles it was the view between the flankers – the highest uprights in the circle. The preference for certain directions has been explained in many ways (Bradley 2016). On individual sites these monuments may have been orientated towards the position of the setting sun, they may have emphasised the position of the moon, or they could command a vista towards a distant hill or mountain (Ruggles and Burl 1985). It is unlikely that one interpretation can account for every example. The chambered cairns at Balnuaran of Clava are directed towards the midwinter sunset, but the recumbent stone and the flankers at Tomnaverie are some way outside the solar arc and, if they did have any celestial alignment (which seems doubtful), it could only have been on the moon (Bradley 2005, 33 and 49–50). On the other hand they are directed towards the summit of Lochnagar 30 km away: an orientation echoed by the neighbouring monument at Hillhead. In that respect they were very different from a variety of Neolithic monuments, especially long cairns, which commonly faced the morning sun.

  The placing of the monoliths and the variations in their heights acted together to establish the axis of individual monuments. That would have been difficult to achieve unless they were unitary constructions. Until all the standing stones were in place their layout would lack any coherence. That is very different from so
me of their predecessors where the stones around their perimeter echoed the form of the surrounding landscape. As Aaron Watson (2001) has demonstrated, a monument like Avebury is situated at the centre of a circular landscape framed by a horizon of hills. That is rarely the case with later constructions. They often face in one direction and emphasise the fact by the grading of the stones. Of course, both principles were acknowledged at particular sites. Thus Tomnaverie seems to be located at the middle of a circular basin, yet the monument is clearly directed towards a mountain on the horizon.

  At the same time there are a few structures at which more subtle relationships were established between the monoliths (Fig. 7.2). They could be paired across opposite sides of the circle according to the colours, textures or shapes of the stones, and individual uprights might also be matched with elements in the kerb of an older cairn (Bradley 2000, chapter 2; 2005, 28–33). In the case of recumbent stone circles there were even subtler patterns in the spacing of the stones (Bradley 2005, illus. 45 and 77). None of these relationships is found especially widely, but, where they do occur, they provide another argument that the buildings had been planned from the outset. They offered less scope for improvisation than the stone circles of the Neolithic period.

  Figure 7.2. Outline plans of Balnuaran of Clava southwest cairn and Tomnaverie summarising the links between the monoliths and the kerbstones according to materials, textures and colours. Information from Bradley 2011.

  Another contrast is more difficult to characterise, but in the case of recumbent stone circles Adam Welfare has observed how often only one kind of rock was used at any site (2011, chapter 5). Whilst there are important exceptions, it is worth observing that, in contrast to some of the monuments considered by Colin Richards, recumbent stone circles make little use of quarried materials or those brought any distance. Often they were glacial erratics. The amount of work required to build a recumbent stone circle does not bear comparison with the task of building Great Stone Circles.